This is one of my play-things: www.p-code.org/ttask
It can be used to compile/launch simulations with GHDL, ModelSim, Xilinx ISim/XSim, and also Icarus Verilog.
There are several small test projects in the test directory (e.g. test/fpga-sim), and with these I also find that GHDL is very reasonable compared to the commercial offerings, and not slow, as was my original expectations as well.
There's also a verilog equivalent project in the test/ directory that I use for Icarus Verilog, and that's pretty good as well.
Also, with regards to simulation speeds, a good way to get good simulation speeds is to have multiple testbenches and run them in parallel. Note that my ttask project has a -threads build property, so if your PC can run 8 hardware threads at a time say, setting -threads to 8 can then give you much reduced simulation times.
I think I read somewhere that Xilinx xsim (Vivado) will only check out one license if you run multiple simulations on one machine, whereas I think Modelsim will check out multiple licences. If that is true, then that is good news for Vivado users. GHDL doesn't have this issue of course!
Regards, Paul. On 06/08/15 19:26, Adrien Prost-Boucle wrote:
Hi everybody, I've launched simulation of more than 12 (personal) VHDL designs, with GHDL, Xilinx's ISim from ISE Design Suite 2013.1 (full) and with Xilinx's XSim from Vivado Design Suite 2015.1 (full). My initial goal was to find how much slower GHDL was compared to these commercial simulators. Note: I compiled GHDL against gcc 4.9 with isl, cloog, lto. Not tested without. My PC is 64 bits ArchLinux. I got a very nice surprise: all simulations were at least 2 times faster, sometimes 4 times faster, than ISim and XSim. For a design with an extreme density of generate statements, with even some recursive component instantiation, XSim did spend 10-15 minutes analyzing the "data flow" during elaboration. The generated simulator did the simulation in ~8 minutes. That design was elaborated in 20 seconds by GHDL, and the simulation took only 3 minutes :-) The only step done slower by GHDL is analysis, because it is not multithreaded. But that has a rather low importance. Anybody can do some comparison with Altera tool or Modelsim? Regards, Adrien _______________________________________________ Ghdl-discuss mailing list [email protected] https://mail.gna.org/listinfo/ghdl-discuss
_______________________________________________ Ghdl-discuss mailing list [email protected] https://mail.gna.org/listinfo/ghdl-discuss
